01
Create a new workflow
Import from a YAML file, copy a previous workflow, or start from scratch with a built-in template.
One platform for every team
For research computing and national labs
Your schedulers, storage, and workflows stay exactly as they are. Fuzzball sits above them and gives every researcher a browser-based way to submit, monitor, and reproduce jobs, so high-performance computing is no longer limited to the people who know the command line.
CIQ trusted by:
FEATURES
Built for the work. Not the setup.
Engineers, researchers, data scientists, and AI teams move from problem to result without touching infrastructure. Administrators control every environment from one place.
Connect to existing storage
Lustre, GPFS, BeeGFS, or any parallel file system. No migration required.
Move data automatically
Pipelines arrive fully provisioned wherever the job runs.
Eliminate external dependencies
A built-in object cache and container registry replace the separate services you would otherwise stand up.
Empower every researcher
A browser interface handles job submission, monitoring, and results. No command line, no infrastructure expertise.
Maximize GPU utilization
Automatic allocation across training and inference keeps GPUs working instead of sitting idle.
Deploy models in one workflow
When training finishes, the inference endpoint provisions itself. No manual handoff.
Keep your schedulers
PBS and Slurm stay in place. Fuzzball schedules containerized, portable workflows through your existing queues.
Run anywhere, unmodified
One workflow definition runs across every cloud and on-premises environment. No rewrites.
CUSTOMER STORY
FYR Diagnostics: 100x the throughput for production-scale genomics
FYR's EV-Omics platform processes complex multi-omics data to identify precision medicine targets. With Fuzzball, their team runs reproducible analytical pipelines at production scale, cutting cycle times from days to hours.
FEDERATE
Run every workload where it performs best
Fuzzball evaluates every available environment and places each job where it runs best, weighing cost, performance, and data locality. Develop in the cloud, run production on-premises, and burst to cloud on demand.
Route jobs intelligently
Each job lands in the environment that fits its cost, performance, and data-locality needs.
Enforce data sovereignty
Placement policies control where regulated workloads run, so models and data stay on infrastructure you control.
Run one definition everywhere
The same workflow runs on every cluster, cloud, and region you operate, without modification.
FUZZBALL IN ACTION
Build and deploy your workflows with ease
Fuzzball's web UI is designed for usability and standardization. Anyone can build, manage, and reuse workflows with just a few clicks:
02
Define your jobs
Parameterize and customize every step of your workflow to match the needs of your jobs.
03
Customize your environment
Specify infrastructure requirements, from instance sizing to storage volumes, ensuring your workflow runs efficiently.
04
Save reusable workflow templates
Workflows can be made into application templates, enabling applications and workflows to be used by anyone.
05
Run, share, and reuse
Fuzzball workflows are modular and portable. Run them on any Fuzzball cluster, chain outputs as inputs to the next workflow, and share templates across your team.
USE CASES
Orchestrate any workload, on any infrastructure.
HPCAI
Unify HPC and AI on one platform
Run simulations and model training on the same cluster from one control plane. No fragmented GPU pools, no separate tools, no duplicate infrastructure.
HPCAI
Burst to cloud when demand spikes
Keep steady-state workloads on-premises and route peak demand to any major cloud automatically, with no change to the job.
HPC
Modernize HPC without rebuilding it
Onboarding drops from weeks to days and cluster utilization climbs, while administrators keep every scheduler, file system, and policy already in production.
AI
Deploy AI models on infrastructure you own
Train and serve models behind production-grade inference endpoints, with data resident in your environment for the full lifecycle.
HPC
Accelerate scientific computing at scale
A browser interface puts high-performance compute in reach of every researcher, so more of the team gets results and fewer jobs wait on specialists.
AI
Meet sovereignty and compliance requirements
Every action is logged and auditable across every environment, which makes regulated workloads straightforward to certify.
WORKS WITH YOUR STACK
Integrate with the tools your team already runs
Works with your applications
Pre-built templates and integrations get common workloads running immediately on familiar tools.
Works with your pipelines
Submit and manage jobs programmatically from existing CI/CD tooling, all the way through to production.
Works with your infrastructure
PBS Professional, Slurm, AWS, S3, Docker, Apptainer, Nextflow, and more. No custom drivers required.
What's new in Fuzzball
A complete self-contained stack
Native storage integration, a built-in object cache, and an integrated container registry are now bundled in, so there are no external services to stand up before work begins.
Multi-cloud deployment
Deploy across AWS, Google Cloud, Oracle Cloud, CoreWeave, and Azure from one control plane.
Production AI inference
Train a model and serve it as a live inference endpoint from a single workflow definition.
RESOURCES
Go deeper on Fuzzball
Tech session
A 45-minute walkthrough of Fuzzball's core capabilities, federation architecture, and the full AI workload lifecycle.
Download the Fuzzball guide.
Everything you need to evaluate Fuzzball for your environment.
Run your first workflow in minutes.
Move data in and out of Fuzzball workflows without an external object store
This is what happens when you don't control your infrastructure
Run Fuzzball workflows on your existing Lustre, GPFS, and BeeGFS storage
CIQ redefines advanced computing with Fuzzball 4.0, democratizing access to HPC for researchers